Subject: Re: i386-current page fault trap in profiled code
To: Frank van der Linden <fvdl@wasabisystems.com>
From: john heasley <heas@shrubbery.net>
List: port-i386
Date: 11/11/2002 14:11:46
Mon, Nov 11, 2002 at 02:59:07PM +0100, Frank van der Linden:
> On Fri, Nov 01, 2002 at 01:27:30PM -0800, john heasley wrote:
> > login: kernel: page fault trap, code=0
> > Stopped in pid 766 (tac_plus) at        fuswintr+0x27:  movzwl           0(%edx)
> > ,%eax
> > db{1}> trace
> > fuswintr(e3335d3c,804ee88,0,e3335d3c) at fuswintr+0x27
> > statclock(e33c4fa4,6,e3242cc8,1000) at statclock+0x82
> > hardclock(e33c4fa4,c03039d0,0,0,4811bd80) at hardclock+0xc9
> > lapic_clockintr(e33c4fa4) at lapic_clockintr+0x73
> > Xintrltimer() at Xintrltimer+0x4a
> 
> This should be fixed now, in sys/arch/i386/i386/trap.c revision 1.174.
> Could you perhaps update your kernel and try again?
> 
> - Frank

seems to occur in a different place now.

login: kernel: page fault trap, code=0
Stopped in pid 28182 (tac_plus) at      0xc010049b:     repe movsb      (%esi),%es:(%edi)
db{1}> trace
gdt_desc(e3f3c014,805fc84,1,1,0) at 0xc010049b
gdt_desc(1f,1f,1f,1f,c) at 0xc01fd012
db{1}> sync
syncing disks... panic: TLB IPI rendezvous failed (mask 1)
Stopped in pid 28182 (tac_plus) at      0xc01ecb34:     movl    %ebp,%esp

i can share the code if you wish.  basically, first thing in main is to
turn off profiling, wait/accept, enable profiling (moncontrol(1)) and
i should get a gmon.out when the child exits.